AI Progress Tracker

    DeepSeek V4.1 Flash · deepseek-v4.1-flash

    DeepSeek · 9 Sep 2026

    Ask the Advisor about this model
    79from 5
    benchmarks

    Not yet confirmed by public reporting. A documented DeepSeek service data exposure concerns infrastructure security and does not establish a weakness in DeepSeek V4.1 Flash.

    context: 1Mtextimage$0.17 / $0.66 per 1M tokens (in / out)

    Benchmark results

    Every figure links to where it was published. Compare like with like — harnesses and prompt setups differ between vendors.

    BenchmarkScoreSource
    GPQA DiamondPhD-level science questions89.8%9 Oct 2026View
    OTIS Mock AIME 2024-25Competition mathematics98.3%8 Oct 2026View
    SimpleQA VerifiedShort factual questions (hallucination resistance)42.9%13 Oct 2026View
    LMArena TextHuman preference Elo from blind chat votes (not a percentage)1464 Elo22 Oct 2026View
    LMArena WebDevHuman preference Elo for building web apps (not a percentage)1665 Elo23 Oct 2026View

    Weighted mean of 3 verified benchmark results: GPQA Diamond 89.8%, OTIS Mock AIME 2024-25 98.3%, SimpleQA Verified 42.9%.

    Security & risk dashboardLatest vulnerabilities, advisories and real-world incidents for DeepSeek V4.1 Flash

    References

    Profile last reviewed 10/27/2026 and re-checked automatically every few hours against the vendor's published documentation, papers and public leaderboards. Scores come from the cited benchmark results above — confirm limits and pricing against the provider's own documentation before building on them.